EMA Environmental Risk Assessment for Generic and Hybrid Applications: Requirements and Precedents for Relying on Existing Data

EMA environmental risk assessment for generic and hybrid applications: requirements and precedents for relying on existing data

Key points

• Generic (Article 10(1)), hybrid (Article 10(3)) and similar biological (Article 10(4)) applications are not exempt from submitting an environmental risk assessment (ERA). The 2024 Revision 1 of EMA's human-medicines ERA guideline says this explicitly ¹¹⁵.
• Submitting an ERA is not the same as submitting new studies. Since the 2006 guideline, EMA has accepted that the absence of ERA studies can be justified for generics, provided the justification addresses any possible significant increase in environmental exposure ¹⁴⁰¹⁴¹.
• Revision 1 adds a generic-specific route. Repeat studies are generally not required where an EU authority has already found a relevant ERA satisfactory and the applicant justifies why its conclusions still apply. Using another company's data, however, still requires written consent or a letter of access ^143505506.
• Published literature can replace or supplement recommended studies if it is reliable (CRED categories 1 or 2) and of comparable design. Public assessment reports and regulatory summaries do not count as ERA dossier data ⁹⁰⁹¹⁹².
• EPAR precedents fall into three groups. Exposure-substitution justifications are routinely accepted for generics. Prevalence- or consumption-refined Phase I calculations and public-domain data have also been accepted. Literature-based arguments without study reports or real data have been rejected; CHMP has then required post-authorisation study commitments, or the application was withdrawn before a conclusion.

Legal basis and regulatory status of the ERA

EMA's guideline describes the obligation under Article 8(3) of Directive 2001/83/EC this way: "the evaluation of the potential environmental risks posed by medicinal products should be submitted, their environmental impact should be assessed and, on a case-by-case basis, specific arrangements to limit it should be envisaged. In any event this impact should not constitute a criterion for refusal of a marketing authorisation" ¹¹⁷. The ERA is required for initial EU marketing authorisation applications regardless of legal basis and is filed in Module 1.6 ^114115116. EMA guidance also states that ERA outcome is not a ground for refusal ¹¹⁷¹¹⁸.

The applicable EMA documents are:

The tiered assessment in brief

• Phase I. The surface-water predicted environmental concentration (PECsw) is calculated as maximum daily dose × market penetration factor (Fpen) ÷ (wastewater per inhabitant per day × dilution factor) ²⁹². The default Fpen, wastewater volume and dilution factor are 0.01, 200 L/inhabitant/day and 10, respectively ²⁹³. Below 0.01 µg/L, with no other concern, risk is considered unlikely. At or above that limit, Phase II follows. The limit may not apply to substances that can affect reproduction at lower concentrations ²⁹⁴.
• PBT/vPvB screening. All active substances are screened regardless of PEC. A log Kow above 4.5 triggers a definitive PBT/vPvB assessment ¹²⁴¹²⁵.
• Phase II. Tier A compares PECsw with the predicted no-effect concentration (PNEC) using physico-chemical, fate and ecotoxicity studies. Where a risk is identified, Tier B refines exposure and, if warranted, effects ¹²³. Under Revision 1, some substance groups enter Phase II irrespective of PEC, and some need tailored testing ¹²².

How EMA applies the ERA to Article 10 applications

No exemption, but studies can be justified away

Under Revision 1 (heading "Q2a: Does the application refer to Article 10 of Directive 2001/83/EC?"), "Applications under Art 10(1)-generic medicinal products, Art 10(3)-hybrid and Art 10(4)-similar biological applications … are not exempt from submitting an ERA". The same applies to Article 10a well-established use, 10b fixed combination and 10c informed consent applications ¹¹⁵.

The 2006 guideline already allowed for this: "There may be cases in which the absence of ERA studies could be justified (e.g. marketing authorisation applications for generic medicinal products or type II variations)". The rationale must take account of any possible significant increase in environmental exposure ¹⁴⁰¹⁴¹. The 2016 Q&A states that suitable evidence of no significant increase in exposure can justify the absence of a complete ERA. It suggests active-ingredient consumption data over time as useful evidence and allows other convincing arguments for a generic ¹⁴⁴¹⁴⁵. The Q&A also confirms that a generic applicant cannot cross-reference the originator's ERA without the originator's consent ²⁸⁹²⁹⁰.

When a generic or hybrid may increase exposure

The Q&A gives these cautions:

• A generic may increase exposure, for example if it is launched in a Member State where the reference product is not marketed ¹⁴⁶.
• For type II variations and line extensions, a new ERA is needed if increased exposure is expected. Examples are a larger patient population, a new indication, a higher maximum recommended dose, or a formulation (such as a patch) that retains significant drug substance after use. Significance is judged case by case ¹⁴⁷.

These triggers matter most for hybrids, which by definition differ from the reference product (for example in strength, dosage form, route or indication).

Revision 1, Q2b: earlier ERAs for the same active substance

Revision 1 sets out how an applicant should handle prior ERA data:

• The ERA is product-specific and remains the applicant's responsibility, even where an earlier ERA exists for the same active substance ⁵⁰⁵.
• Data sharing is encouraged, particularly to avoid animal studies. Another MAH's study reports may be submitted with the data owner's written consent ⁵⁰⁵.
• If sharing is not agreed, the applicant must still provide a complete ERA. Specifically for generics, repeating studies will generally not be required where an EU national competent authority found a relevant ERA satisfactory and the applicant justifies why its scientific conclusions apply to the generic. The generic's SmPC and package leaflet should then carry information and risk-mitigation measures aligned with that ERA ¹⁴³. The Q2b passage does not expressly extend this route to hybrids ¹⁴³.
• If the earlier ERA is incomplete against current standards (missing studies, or possible increased exposure), the applicant should perform the missing studies and/or update the ERA ⁵⁰⁷⁵⁰⁸.
• Where risk cannot be excluded, or the substance is PBT/vPvB, the SmPC should reflect this and the applicant should propose measures to limit environmental impact ⁵⁰⁹⁵¹⁰.

Naturally occurring substances

Vitamins, electrolytes, amino acids, peptides, proteins, nucleotides, carbohydrates and lipids are not exempt. However, their ERA may consist of a justification for not submitting studies, for example on physicochemical or environmental-fate grounds. Herbal products are treated in a similar way ^120142121.

Using published literature and existing data instead of new studies

Revision 1's sections on publicly available data and on evaluation of studies set the conditions:

• For an already marketed active substance, the applicant should provide a complete literature review so that relevant existing data are used and animal studies are not repeated. Study reports and peer-reviewed literature must contain enough detail to assess reliability ⁸⁹.
• Published studies of acceptable quality may replace or supplement recommended experimental studies, support a tailored assessment, or update an ERA obtained from another MAH. The applicant should document the search, list relevant studies with their reliability scores, and make sure that any literature study used in place of a recommended study has a comparable design. Published studies need not be GLP-compliant ⁹⁰.
• Reliability and relevance should be assessed consistently using a standardised method such as CRED. Only CRED categories 1 and 2 may be used in the risk assessment. Categories 3 and 4 may not ⁹¹.
• PARs, EPARs and reviews or summary data from other regulatory frameworks are not ERA dossier data in themselves. Using another company's endpoints requires a letter of access, and the applicant should hold and submit the underlying study reports. Older endpoints, or endpoints assessed under different standards, may not meet current requirements ⁹². A prior assessment under REACH or another framework therefore does not by itself exempt a substance from fate or ecotoxicity testing ⁹²⁹³.

EPAR precedents

The examples below come from EMA EPARs. Unless an EPAR states the legal basis, the application type is given only as far as the cited passage supports it.

1. Exposure-substitution justification accepted (no ERA studies)

This is the most common approach for centrally authorised generics. The standard wording is that the product "is considered unlikely to result in any significant increase in the combined sales volumes" and environmental exposure, so "the ERA is expected to be similar".

The Kigabeq commitment applies the Q&A caution about launches in Member States without the reference product ¹⁴⁶.

2. Refined Phase I using consumption or prevalence data

Some generic and hybrid applicants calculated a Phase I PEC instead of relying only on a substitution argument. They used real consumption or prevalence data to refine Fpen.

• Trabectedin Accord (generic): European consumption data and published epidemiology refined exposure. PECsw fell below the action limit, and there was no non-clinical objection ²²⁵²²⁶.
• Teriflunomide Viatris (generic): Annual consumption and European prevalence data supported Phase I. Phase II was unnecessary and the ERA was acceptable ²²⁷²²⁸.
• Nordimet (methotrexate, hybrid): A prevalence-based Phase I gave a PECsw below 0.01 µg/L, so Phase II was not needed ⁵⁷⁹⁵⁶⁵.
• Puldysa (idebenone, hybrid): The applicant was asked to recalculate PEC using prevalence for the current indication and 100% market share. The revised PEC was below 0.01 µg/L and drew on environmental data from the applicant's earlier dossier ⁵⁷⁷⁵⁷⁸.
• Teriparatide Sun (hybrid): Phase I PECsw of 0.0001 µg/L. No further assessment was needed ^567568569.

3. Public-domain and literature data accepted

• Fulvestrant Mylan (fulvestrant): Public-domain physicochemical, bioconcentration, aquatic toxicity and environmental-fate data, together with known metabolism, supported the case that Phase II was not required. This held despite fulvestrant's potential as an endocrine disruptor ²²¹⁴⁹⁰. The ERA was judged acceptable ⁴⁹¹. The cited passages do not state the legal basis ⁴⁸⁹.
• Tuzulby (methylphenidate, hybrid): PEC exceeded the action limit in some countries. Published information supported a conclusion that no environmental risk was expected, although no ecotoxicity data were available to derive a PNEC. The applicant was reminded to complete the ERA under the revised guideline, and the product was considered approvable ⁵⁵³⁵⁷⁴.
• Doxorubicin Hydrochloride Tillomed (hybrid): A PEC of 0.45 µg/L triggered Phase II. A published Daphnia magna-derived PNEC gave a PEC/PNEC below 1 ²²³²²⁴. The application was withdrawn, and no CHMP ERA conclusion is recorded ²⁰¹.
• Sancuso (granisetron, hybrid): The MAH was recommended to perform a literature-based Phase II analysis covering patch use and disposal ⁵⁷². This is consistent with the Q&A's reference to patches that retain drug substance ¹⁴⁷.

4. Literature or justification not accepted

These cases show where CHMP has drawn the line:

• Imatinib Koanaa (Article 10(3) hybrid): PECsw was 4.275 µg/L, well above 0.01 µg/L ²⁷⁰⁵³⁹. The applicant argued that sales would be redistributed among imatinib products. CHMP found the argument unsupported by actual data and did not accept it. It requested a full Phase II and an experimental log Kow study report ⁵⁴⁰⁵⁴¹. A literature log Kow value had been supplied without a study report ⁵⁷⁰. There was no non-clinical objection, subject to a post-authorisation commitment to submit the ERA studies ⁵⁴².
• Efmody (hydrocortisone): The applicant cited literature on ready biodegradability to avoid tailored Phase II studies. CHMP did not accept the published reference ⁵¹⁵. It recommended a ready biodegradability test and the protocol or literature supporting the log Kow value post-authorisation ⁵¹⁶. CHMP did accept reliance on literature for non-clinical safety ⁵¹⁴.
• Pravafenix (fenofibrate/pravastatin): An ERA based on bibliographic references for both substances was considered not yet acceptable ²³⁰.
• Diractin (ketoprofen): An ERA based on published data was incomplete and could not rule out risk or the need for Phase II. The application was withdrawn ²⁴⁶¹⁹⁹.
• Qtrilmet (metformin/saxagliptin/dapagliflozin): Existing single-substance ERAs and a published review of metformin fate were reused. Newer literature on possible effects on fish reproduction led CHMP to recommend a fish reproduction study post-approval ²³¹.
• Taxotere (docetaxel, indication extension): Fate and effects information was supplied without study reports, along with literature log Kow values. CHMP found this inadequate, and the MAH made commitments on bioaccumulation testing and a Phase II ERA ²⁴⁴²⁴⁵.

Practical lessons for generic and hybrid applicants

1. Always file Module 1.6. A justification is an ERA. Leaving the section empty is not ¹¹⁵¹⁴⁴.
2. Support substitution claims with data. Standard sales-volume wording has been widely accepted for straightforward generics ⁴⁰⁵³⁶¹. Imatinib Koanaa shows CHMP can reject the argument when PEC is high and no real data back the claim ⁵⁴⁰⁵⁴¹.
3. Check hybrid differences against the Q&A triggers: new indication or population, higher maximum dose, new Member State markets, and formulations such as patches ¹⁴⁶¹⁴⁷. Refined Phase I calculations using prevalence or consumption data have been an effective middle path ^227579577.
4. Qualify literature properly. Score reliability (CRED 1 or 2), show comparable study design, and do not rely on EPARs or regulatory summaries as data ⁹⁰⁹¹⁹². Literature log Kow values without study reports have repeatedly been challenged ^570516244.
5. Plan for data access early. Cross-referencing another company's ERA requires consent or a letter of access ²⁸⁹⁵⁰⁶. Without it, the Revision 1 Q2b route for generics depends on a satisfactory prior EU assessment and a scientific bridging justification ¹⁴³.
6. Expect post-authorisation commitments rather than refusal. ERA is not a refusal criterion ¹¹⁷. Gaps are normally closed through commitments to submit studies after approval ^542516231.